Two key characteristics of the neo-Darwinian synthesis in evolutionary biology have been its emphasis on the importance of mutations of small effect (micromutationism) and the view that studies of individual gene function shed relatively little light on evolutionary processes. Recent advances in molecular biology, however, have broken down many of the barriers between functional and evolutionary inquiry, opening the door to detailed studies of the genetic basis of functional trait evolution in plants. In this article, we review the insights into plant evolution that have been provided by molecular methods and address future research needs. Quantitative trait locus (QTL) mapping in crop and model plants has shown that individual loci often have large effects on trait variation, at variance with the micromutationist perspective. Evidence so far indicates that QTLs with large effects are also important in wild populations, underlying interspecific differences as well as intraspecific variation. Isolation of some of these QTLs, in particular for flowering time variation, has revealed a prominent role for regulatory genes known to function in regulation of flowering and exposed the complexity of regulatory processes. Preliminary evidence indicates that plant growth variation may be directly regulated rather than primarily the indirect result of selection on constituent processes. Future research should expand the number of traits that are intensively studied and make greater use of QTL mapping in wild plant taxa, especially those undergoing adaptive radiations, while continuing to draw on insights from model plants. Promising techniques include testing of candidate gene-trait associations in wild populations, genetic mapping in hybrid zones, and microarray analyses of gene expression.